Performance evaluation system for auxiliary driving system

Abstract

The invention discloses an auxiliary driving system performance evaluation system, and relates to the technical field of system performance evaluation.The auxiliary driving system performance evaluation system comprises a driving parameter acquisition module, a test grade confirmation module, a performance test module, a performance evaluation module and a data bin. The testing accuracy of the automatic driving trolley is improved, the waste of testing resources is reduced, and the sustainable development of the automatic driving trolley is guaranteed; according to the method, the stability of the auxiliary driving system of the automatic driving trolley is analyzed, so that the avoidance stability of the auxiliary driving system of the automatic driving trolley is guaranteed, the stable operation of hardware equipment of the automatic driving trolley is maintained while the oil consumption or electric quantity loss of the automatic driving trolley is reduced, and the reliability of the automatic driving trolley is improved. Therefore, the service life of the self-driving trolley can be maintained.

Description

Technical Field

[0001] The present invention relates to the technical field of system performance evaluation, and particularly to a performance evaluation system for an assisted driving system. Background Art

[0002] In the current rapid development of technology, autonomous vehicles, especially driverless taxis, are gradually moving from science fiction concepts into people's daily lives. They carry high hopes for improving traffic efficiency, reducing human driving errors, and revolutionizing the way of travel. The assisted driving performance of driverless taxis not only concerns the safety of passengers' lives but also affects the healthy development of the entire autonomous driving industry. Therefore, it is of great importance and necessity to conduct a comprehensive, in-depth, and continuous evaluation of its assisted driving performance.

[0003] The prior art, such as an invention application patent with the publication number CN109435955B, discloses a method, device, equipment, and storage medium for evaluating the performance of an autonomous driving system. The method includes: obtaining scenario information, where the scenario information includes the source data used for evaluating the performance of each functional module of the autonomous driving system; for each functional module, based on the source data corresponding to the current functional module and the output data of the upper-level functional module in the evaluation process, evaluating the performance of the current functional module to obtain a sub-functional performance evaluation result corresponding to the current functional module; obtaining the overall performance evaluation result of the autonomous driving system based on the obtained sub-functional performance evaluation results; where the function implementation of the current functional module depends on the output data of the upper-level functional module in the evaluation process. Through the above technical solution, the evaluation of the overall performance of the autonomous driving system is achieved, and the overall ability of the autonomous driving system can be known according to the evaluation result.

[0004] The prior art, such as an invention application patent with the publication number CN114896754A, discloses a method for evaluating the performance of an autonomous driving system for the full parameter space of a logical scenario. After a given test logical scenario of the autonomous driving system and its matching parameter space are provided, the autonomous driving system to be tested is placed in this logical scenario for testing to obtain the driving data under each specific test condition. After obtaining the driving trajectory field of the autonomous driving system to be tested during the test process in the entire parameter space of the tested logical scenario, first, the logical scenario is partitioned into a safe area and a dangerous area according to the ideal vehicle motion curve; then, the evaluation focuses and evaluation indicators in the two partitions are determined; and the overall performance in the entire parameter space is calculated based on the test results to obtain the performance evaluation result of the entire full parameter space.

[0005] Combining the above solutions, it can be found that in the prior art, there are few performance evaluation parameters for the assisted driving system based on autonomous driving vehicles to determine the test level of autonomous driving vehicles. The personalization of the test for autonomous driving vehicles is weak, which reduces the accuracy of the test for autonomous driving vehicles, increases the waste of test resources to a certain extent, affects the sustainable development of autonomous driving vehicles, and in the evaluation of the performance evaluation parameters of the assisted driving system of the driving vehicle, the attention to the stability of the assisted driving system of the autonomous driving vehicle in the case of avoidance and poor road conditions is not high, so it is difficult to ensure the avoidance stability of the assisted driving system of the autonomous driving vehicle, increases the fuel consumption or power loss of the autonomous driving vehicle, and has an adverse impact on the hardware equipment of the autonomous driving vehicle, thereby reducing the lifespan of the autonomous driving vehicle. Summary of the Invention

[0006] The purpose of the present invention is to provide a performance evaluation system for an assisted driving system, which solves the problems existing in the background technology.

[0007] To solve the above technical problems, the present invention adopts the following technical solutions: The present invention provides a performance evaluation system for an assisted driving system, including: a driving parameter acquisition module for acquiring the driving parameters of the assisted driving systems of each dispatched vehicle;

[0008] A test level confirmation module for generating a set of test levels for each dispatched vehicle based on the driving parameters of the assisted driving systems of each dispatched vehicle and combining the meteorological parameters of the assisted driving systems of each dispatched vehicle;

[0009] A performance test module for performing performance tests on the assisted driving systems of each dispatched vehicle in the vehicle operation and maintenance management center based on the set of test levels for each dispatched vehicle to obtain the test data of the assisted driving systems of each dispatched vehicle;

[0010] A performance evaluation module for judging whether a dispatched vehicle needs to be recalled based on the test data of the assisted driving system of each dispatched vehicle and sending the numbers of the vehicles that need to be recalled to the vehicle operation and maintenance management center.

[0011] The beneficial effects of the present invention are as follows: (1) Based on the performance evaluation parameters of the assisted driving system of the autonomous driving vehicle, the present invention determines the test level of the autonomous driving vehicle, has a high degree of personalization in the test of the autonomous driving vehicle, improves the accuracy of the test of the autonomous driving vehicle, reduces the waste of test resources to a certain extent, and ensures the sustainable development of the autonomous driving vehicle.

[0012] (2) In the evaluation of the performance evaluation parameters of the assisted driving system for a driving car, the present invention analyzes the stability of the assisted driving system of the autonomous driving car in the case of avoidance and poor road conditions, so as to ensure the avoidance stability of the assisted driving system of the autonomous driving car, reduce the fuel consumption or power loss of the autonomous driving car, and maintain the stable operation of the hardware equipment of the autonomous driving car, thereby contributing to maintaining the lifespan of the autonomous driving car. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0014] Figure 1 It is a schematic diagram of the system structure connection of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0015] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0016] Refer to Figure 1 As shown, the present invention provides a performance evaluation system for an assisted driving system, including: a driving parameter acquisition module, a test level confirmation module, a performance test module, a performance evaluation module, and a data warehouse.

[0017] It should be noted that the driving parameter acquisition module is connected to the test level confirmation module, the test level confirmation module is connected to the performance test module, the performance test module is connected to the performance evaluation module, and the data warehouse is respectively connected to the test level confirmation module, the performance test module, and the performance evaluation module.

[0018] The driving parameter acquisition module is used to obtain the driving parameters of the assisted driving system of each dispatched vehicle from the vehicle operation and maintenance management center;

[0019] In a specific embodiment of the present invention, the driving parameters include the trajectories, speed data sets, recognition and avoidance data sets, and road condition data sets of each driving;

[0020] The meteorological parameters include the characteristic parameters of each meteorological data of each driving.

[0021] It should be noted that the characteristic parameters of the meteorological data include, but are not limited to, the maximum temperature, the maximum humidity, the temperature difference, and the humidity difference.

[0022] The test level confirmation module is used to generate a set of test levels for each dispatched vehicle based on the driving parameters of the assisted driving system of each dispatched vehicle and in combination with the meteorological parameters of the assisted driving system of each dispatched vehicle.

[0023] In a specific embodiment of the present invention, the method for generating the set of test levels for each dispatched vehicle is as follows: Based on the driving parameters and meteorological data of the assisted driving system of each dispatched vehicle, evaluate the performance evaluation parameter α(PI) of the assisted driving system of each dispatched vehicle _i , where i is the number of each dispatched vehicle, i = 1, 2,..., n;

[0024] If then this dispatched vehicle is taken as an element in the first test level set of dispatched vehicles, is the performance evaluation parameter interval corresponding to the first test level of the dispatched vehicle stored in the data warehouse;

[0025] If is the performance evaluation parameter interval corresponding to the second test level of the dispatched vehicle stored in the data warehouse;

[0026] If is the performance evaluation parameter interval corresponding to the third test level of the dispatched vehicle stored in the data warehouse;

[0027] Summarize several dispatched vehicles in the first test level set of dispatched vehicles to construct the first test level set of dispatched vehicles where are the 1st dispatched vehicle, the 2nd dispatched vehicle, the 3rd dispatched vehicle, and the lth dispatched vehicle respectively. Similarly, summarize several dispatched vehicles in the second test level set of dispatched vehicles to construct the second test level set of dispatched vehicles, and summarize several dispatched vehicles in the third test level set of dispatched vehicles to construct the third test level set of dispatched vehicles;

[0028] Based on the first test level set, the second test level set, and the third test level set of dispatched vehicles, summarize to obtain the set of test levels for each dispatched vehicle.

[0029] In a specific embodiment of the present invention, the method for specifically evaluating the performance evaluation parameter of the assisted driving system of each dispatched vehicle is as follows: Obtain the trajectory, speed data set, recognition and avoidance data set, and road condition data set of each driving from the driving parameters of each dispatched vehicle, and process to obtain the avoidance performance evaluation parameter and the conventional performance evaluation parameter of each dispatched vehicle during each driving;

[0030] The speed data set includes the speeds at various time points;

[0031] The recognition and avoidance data set includes the obstacle information, start time point, end time point, and risk characteristic parameters for each avoidance;

[0032] It should be noted that the risk characteristic parameter is specifically a numerical value of 0 or 1. If an avoidance is successful, the risk characteristic parameter is 0; otherwise, the risk characteristic parameter is 1.

[0033] The road condition data set includes the set of road condition images and the set of distances between the chassis and the road surface for each detection period;

[0034] The characteristic parameters of each meteorological data are obtained from the meteorological parameters of each dispatched vehicle, and the meteorological evaluation parameters for each driving of each dispatched vehicle are obtained through data processing;

[0035] The avoidance performance evaluation parameter ε of each dispatched vehicle for each driving _i 、the conventional performance evaluation parameter η _i and the meteorological evaluation parameter β _i are imported into the performance evaluation parameter model of the assisted driving system to output the performance evaluation parameter α(PI) of the assisted driving system of each dispatched vehicle _i , where ε′, η′, β′, ε _i→1 、η _i→1 are respectively the ideal avoidance performance evaluation parameter, ideal conventional performance evaluation parameter, ideal meteorological evaluation parameter, and allowable deviation avoidance performance evaluation parameter corresponding to the unit deviation meteorological evaluation parameter, and the allowable deviation conventional performance evaluation parameter in the data warehouse.

[0036] In a specific embodiment of the present invention, for the avoidance performance evaluation parameter of each dispatched vehicle for each driving, the specific evaluation method is as follows: The start time point, end time point, and risk characteristic parameter F of each avoidance are extracted from the recognition and avoidance data set of each driving of each dispatched vehicle _imp , m is the number of each driving, m = 1, 2,..., l, and p is the number of each avoidance, p = 1, 2,..., q;

[0037] Based on the start time point and end time point of each avoidance of each driving of each dispatched vehicle, combined with the trajectory of each driving of each dispatched vehicle, the trajectory of each avoidance of each driving of each dispatched vehicle is intercepted, and the speed data subset of each avoidance of each driving of each dispatched vehicle is intercepted. After data processing, the first avoidance performance evaluation parameter ε1 of each driving of each dispatched vehicle is output _i ;

[0038] Extract the road surface condition image set and the chassis-to-road surface distance set for each monitoring period from the road condition data sets of each trip of each dispatched vehicle, determine whether the road surface condition of each monitoring period of each trip of each dispatched vehicle is abnormal. If the road surface condition of a certain monitoring period is abnormal, record this monitoring period as an abnormal monitoring period, and filter out the abnormal monitoring periods of each trip of each dispatched vehicle;

[0039] It should be noted that the specific method for determining whether the road surface condition of each monitoring period of each trip of each dispatched vehicle is abnormal is as follows: Based on the road surface condition image set of each monitoring period of each trip of each dispatched vehicle, identify the road surface defect parameters through image recognition technology, where the road surface defect parameters include each defect type, and each defect type includes but is not limited to wet and slippery, concave, and convex;

[0040] Based on the chassis-to-road surface set of each monitoring period of each trip of each dispatched vehicle, evaluate the data variance in the chassis-to-road surface set of each monitoring period of each trip of each dispatched vehicle. The calculation method of the variance is prior art and will not be elaborated here;

[0041] When there is a defect type in a certain monitoring period or the data variance in the chassis-to-road surface set is greater than or equal to the preset data variance threshold, it is determined that the road surface condition of this monitoring period is abnormal. When there is no defect type in a certain monitoring period and the data variance in the chassis-to-road surface set is less than the preset data variance threshold, it is determined that the road surface condition of this monitoring period is normal.

[0042] Based on the abnormal monitoring periods of each trip of each dispatched vehicle, intercept the trajectory and speed data subsets corresponding to the abnormal monitoring periods of each trip of each dispatched vehicle. After data processing, output the second avoidance performance evaluation parameter ε2 of each trip of each dispatched vehicle _i ;

[0043] It should be noted that, consistent with the evaluation method of the first avoidance performance evaluation parameter of each trip of each dispatched vehicle, evaluate the second avoidance performance evaluation parameter of each trip of each dispatched vehicle.

[0044] Import the first avoidance performance evaluation parameter and the second avoidance performance evaluation parameter of each trip of each dispatched vehicle into the avoidance performance evaluation parameter evaluation model ε _i =ln(1 + ε1 _i *λ1 + ε2 _i *λ2), and output the avoidance performance evaluation parameter of each trip of each dispatched vehicle. In the formula, λ1 and λ2 respectively represent the weight influence factors corresponding to the preset first avoidance performance evaluation parameter and the second avoidance performance evaluation parameter.

[0045] It should be noted that the weight influence factors corresponding to the first avoidance performance evaluation parameter and the second avoidance performance evaluation parameter are specifically set and uploaded by the testers of driverless vehicles.

[0046] In a specific embodiment of the present invention, the specific processing process of the first avoidance performance evaluation parameter for each driving of each dispatching vehicle is as follows: through the trajectories of each avoidance in each driving of each dispatching vehicle, obtain the convergence distance, trajectory length, and avoidance duration of each avoidance in each driving of each dispatching vehicle, and evaluate the trajectory smoothness index of each avoidance in each driving of each dispatching vehicle. In the formula represents the j-th vector of the p-th avoidance of the m-th driving of the i-th dispatching vehicle;

[0047] It should be noted that the convergence distance is specifically the closest distance between the dispatching vehicle and the obstacle. Each vector of each avoidance in each driving of each dispatching vehicle is specifically obtained by dividing the trajectory of each avoidance in each driving of each dispatching vehicle into several line segments according to a preset segmentation length, and obtaining the three-dimensional coordinates of the starting point and the ending point of the several line segments, and calculating the direction vectors of the several line segments. The calculation formula of the direction vector is prior art and will not be elaborated here.

[0048] It should also be noted that Specifically, it is obtained by calculating the dot product of two vectors divided by the product of their respective magnitudes;

[0049] Based on the obstacle information of each avoidance in each driving of each dispatching vehicle and the reference avoidance information stored in the data warehouse, where the reference avoidance information includes the basic information of each obstacle and the avoidance trajectory, determine the reference convergence distance, reference trajectory smoothness index, trajectory length, and avoidance duration of each avoidance in each driving of each dispatching vehicle.

[0050] It should be noted that the specific determination method of the reference convergence distance, reference trajectory smoothness index, trajectory length, and avoidance duration of each avoidance in each driving of each dispatching vehicle is as follows: match the obstacle information of each avoidance in each driving of each dispatching vehicle with the basic information of each obstacle one by one. If the obstacle information of a certain avoidance is consistent with the basic information of a certain obstacle, obtain the avoidance trajectory of the obstacle, and obtain the convergence distance, trajectory smoothness index, trajectory length, and avoidance duration as the reference convergence distance, reference trajectory smoothness index, trajectory length, and avoidance duration of this avoidance, so as to determine the reference convergence distance, reference trajectory smoothness index, trajectory length, and avoidance duration of each avoidance in each driving of each dispatching vehicle.

[0051] It should also be noted that the obstacle information includes but is not limited to obstacle volume, obstacle contour, obstacle length, and obstacle width, etc.

[0052] Determine the speed dispersion index V of each avoidance during each trip of each dispatched vehicle through the subset of speed data of each avoidance during each trip of each dispatched vehicle _imp ;

[0053] It should be noted that the specific method for determining the speed dispersion index of each avoidance during each trip of each dispatched vehicle is as follows: Based on the subset of speed data of each avoidance during each trip of each dispatched vehicle, evaluate the data variance in the subset of speed data of each avoidance during each trip of each dispatched vehicle, and use it as the speed dispersion index of each avoidance during each trip of each dispatched vehicle. The calculation method of the variance is prior art and will not be elaborated here.

[0054] Pass through the first avoidance performance evaluation parameter model Output the first avoidance performance evaluation parameter χ of each trip of each dispatched vehicle _imph , χ' _imph Are respectively the h-th data and the h-th reference data of the p-th avoidance of the m-th trip of the i-th dispatched vehicle, h ∈ [1, 4].[[]END]

[0055] It should be noted that the data of each avoidance includes the convergence distance, the trajectory smoothness index, the trajectory length, and the avoidance duration, and the reference data includes the reference convergence distance, the reference trajectory smoothness index, the trajectory length, and the avoidance duration.

[0056] In a specific embodiment of the present invention, the specific evaluation method for the conventional performance evaluation parameter of each dispatched vehicle during each trip is as follows: Based on the trajectories of each avoidance during each trip of each dispatched vehicle and the trajectories of each abnormal monitoring period intercepted, obtain each section of the conventional driving trajectory of each dispatched vehicle during each trip, and obtain the reference driving trajectory of each dispatched vehicle during each trip from the data warehouse. After trajectory fitting processing, obtain the trajectory accuracy index Q_im of each dispatched vehicle during each trip;

[0057] It should be noted that the specific method for obtaining the accuracy index of each section of the conventional driving of each dispatched vehicle during each trip is as follows: Compare each section of the conventional driving trajectory of each dispatched vehicle during each trip with the reference driving trajectory, obtain the overlapping length of the conventional driving trajectory and the reference driving trajectory of each dispatched vehicle during each trip, and obtain the length of the reference driving trajectory of each dispatched vehicle during each trip. Divide the overlapping length of the conventional driving trajectory and the reference driving trajectory of each dispatched vehicle during each trip by the length of the reference driving trajectory to obtain the trajectory accuracy index of each dispatched vehicle during each trip;

[0058] It should also be noted that the reference driving trajectories of each dispatched vehicle during each trip are specifically the predicted driving trajectories of each dispatched vehicle during each trip after intercepting the corresponding time periods from the predicted driving trajectories of each dispatched vehicle during each trip according to the start time point, end time point of each avoidance, and each abnormal monitoring period of each dispatched vehicle during each trip, and the predicted driving trajectories of each dispatched vehicle during each trip are specifically the paths automatically planned by the dispatched vehicle based on the assisted driving system.

[0059] Based on the speed data subsets of each avoidance and the speed data subsets of each abnormal monitoring period obtained by intercepting during each trip of each dispatched vehicle, the speed data subsets of each section of normal driving of each dispatched vehicle during each trip are obtained, and the speed dispersion index L of each section of normal driving of each dispatched vehicle during each trip is determined. _imf , where f is the number of each section of normal driving, f = 1, 2,..., t;

[0060] It should be noted that, in the same way as the method for confirming the speed dispersion index of each avoidance of each dispatched vehicle during each trip, the speed dispersion index of each section of normal driving of each dispatched vehicle during each trip is determined.

[0061] Through the normal performance evaluation parameter model Output the normal performance evaluation parameters of each dispatched vehicle during each trip, where e is the natural constant.

[0062] In a specific embodiment of the present invention, for the meteorological evaluation parameters of each dispatched vehicle during each trip, the specific evaluation method is: importing the characteristic parameters of each meteorological data of each trip in the meteorological parameters of each dispatched vehicle into the meteorological evaluation parameter model to output the meteorological evaluation parameters of each dispatched vehicle during each trip, where δ _imb is the characteristic parameter of the b-th meteorological data of the m-th trip in the meteorological parameters of the i-th dispatched vehicle, and δ _ ′ imb is the characteristic parameter of the b-th meteorological data in the ideal meteorological parameters stored in the data warehouse, and b is the number of each meteorological data, b = 1, 2,..., d.

[0063] In the evaluation of the performance evaluation parameters of the assisted driving system of the driving car in the present invention, the stability of the assisted driving system of the autonomous driving car in the case of avoidance and poor road conditions is analyzed, so as to ensure the avoidance stability of the assisted driving system of the autonomous driving car, reduce the fuel consumption or power loss of the autonomous driving car, and maintain the stable operation of the hardware equipment of the autonomous driving car, thereby being beneficial to maintaining the lifespan of the autonomous driving car.

[0064] Based on the performance evaluation parameters of the auxiliary driving system of the autonomous driving vehicle, the test level of the autonomous driving vehicle is determined, which is highly personalized for the test of the autonomous driving vehicle, improves the accuracy of the test of the autonomous driving vehicle, reduces the waste of test resources to a certain extent, and ensures the sustainable development of the autonomous driving vehicle.

[0065] The performance test module is used to perform performance tests on the auxiliary driving systems of each dispatched vehicle in the vehicle operation and maintenance management center based on the set of test levels of each dispatched vehicle, and obtain the test data of the auxiliary driving systems of each dispatched vehicle.

[0066] In a specific embodiment of the present invention, the method for performing performance tests on the auxiliary driving systems of each dispatched vehicle is as follows: based on the set of test levels of each dispatched vehicle, and obtaining the interval step lengths of each test level in its respective variable test items from the data warehouse, screening the interval step lengths of the set of test levels of each dispatched vehicle in its respective variable test items, and using the control variable method to set the characteristic values of each test of the set of test levels of each dispatched vehicle in its respective variable test items.

[0067] In a specific embodiment, if the variable test item is temperature and the interval step length of the set of test levels of each dispatched vehicle in the temperature test item is R, then the lowest temperature in the extreme temperature range in the data warehouse is used as the characteristic value of the first test, and it is added to R to obtain the characteristic value of the second test, and the characteristic value of the second test is added to R to obtain the characteristic value of the third test. If the characteristic value of a certain test is less than the highest temperature in the extreme temperature range and the characteristic value of the next test of this test is greater than the highest temperature in the extreme temperature range, then the highest temperature in the extreme temperature range is recorded as the characteristic value of the next test of this test, and the characteristic values of each test of the set of test levels of each dispatched vehicle in the temperature test item are obtained.

[0068] If the variable test item is road condition and the interval step length of the set of test levels of each dispatched vehicle in the road condition test item is U, then in combination with the extreme road condition quality coefficient range in the data warehouse, similarly, the characteristic values of each test of the set of test levels of each dispatched vehicle in the road condition test item are confirmed.

[0069] Based on the characteristic values of each test of the set of test levels of each dispatched vehicle in its respective variable test items, performance tests are performed on the auxiliary driving systems of each dispatched vehicle.

[0070] The test data includes the performance evaluation parameters of each test of its respective variable test items.

[0071] It should be noted that the analysis methods of the performance evaluation parameters of each test of each dispatched vehicle in its respective variable test items are the same as those of the performance evaluation parameters of the auxiliary driving system of each dispatched vehicle.

[0072] The performance evaluation module is used to determine whether a dispatched vehicle needs to be recalled based on the test data of the assisted driving systems of the dispatched vehicles, and send the numbers of the vehicles that need to be recalled to the vehicle operation and maintenance management center.

[0073] In a specific embodiment of the present invention, the method for determining whether a dispatched vehicle needs to be recalled is as follows: obtain the performance evaluation parameters of each test of each independent test item from the test data of the assisted driving systems of the dispatched vehicles. If the performance evaluation parameter of a certain test is greater than or equal to the preset performance evaluation parameter threshold, then record this test as a successful test, and count the number of successful tests of each independent test item of the assisted driving systems of the dispatched vehicles and the total number of tests and import them into the test performance evaluation index model to output the test performance evaluation indexes of the assisted driving systems of the dispatched vehicles. In the formula is the number of successful tests of a certain independent test item of the i-th dispatched vehicle, τ′ _x is the successful test ratio of the x-th independent item stored in the data warehouse, x is the number of each independent test item, x = 1, 2,..., y;

[0074] The test performance evaluation index includes values of 0 and 1. When the test performance evaluation index is 0, it indicates that the test performance of the assisted driving system of the dispatched vehicle is unqualified and the dispatched vehicle needs to be recalled. When the test performance evaluation index is 1, it indicates that the test performance of the assisted driving system of the dispatched vehicle is qualified and the dispatched vehicle does not need to be recalled.

[0075] It should be understood that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0076] The above content is only an example and illustration of the concept of the present invention. Those skilled in the art of the present technology can make various modifications or supplements to the described specific embodiments or use similar methods for substitution, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should all belong to the protection scope of the present invention.

Claims

1. A driving assistance system performance evaluation system, characterized in that: include: A driving parameter acquisition module is used to obtain the driving parameters of the auxiliary driving system of each dispatched vehicle; A test level confirmation module, used to generate a set of test levels for dispatched vehicles based on driving parameters of the assisted driving system of each dispatched vehicle and in combination with meteorological parameters of the assisted driving system of each dispatched vehicle; The performance test module is used to perform the assisted driving system performance test on each dispatched vehicle in the vehicle operation and maintenance management center based on the test level sets of the dispatched vehicles, and obtain the test data of the assisted driving system of each dispatched vehicle; The performance evaluation module is used to determine whether the dispatched vehicles need to be recalled based on the test data of the assisted driving system of each dispatched vehicle, and send the numbers of the vehicles that need to be recalled to the vehicle operation and maintenance management center.

2. The driver assistance system performance evaluation system according to claim 1, characterized in that: The driving parameters include the trajectory of each driving, the speed data set, the identification and avoidance data set and the road condition data set; The meteorological parameters include characteristic variables of each meteorological data of each driving.

3. The driving assistance system performance evaluation system according to claim 2, characterized in that: The specific generation method of generating the test level sets of dispatched vehicles is as follows: Based on the driving parameters and weather data of the assisted driving system of each dispatched vehicle, the performance evaluation parameter α(PI) of the assisted driving system of each dispatched vehicle is evaluated. _i , i is the number of each dispatched vehicle, i = 1, 2, ..., n; like The dispatched vehicle is taken as an element in the first test level set of dispatched vehicles. is the performance evaluation parameter interval corresponding to the first test level of the dispatched vehicle stored in the data warehouse; like is the performance evaluation parameter interval corresponding to the second test level of the dispatched vehicle stored in the data warehouse; like is the performance evaluation parameter interval corresponding to the third test level of the dispatched vehicle stored in the data warehouse; Aggregate several dispatched vehicles in the dispatched vehicle first test level set to construct the dispatched vehicle first test level set in They are the first dispatched vehicle, the second dispatched vehicle, the third dispatched vehicle and the lth dispatched vehicle respectively. Similarly, several dispatched vehicles in the second test level set of dispatched vehicles are aggregated to construct the second test level set of dispatched vehicles, and several dispatched vehicles in the third test level set of dispatched vehicles are aggregated to construct the third test level set of dispatched vehicles; According to the first test level set, the second test level set and the third test level set of the dispatched vehicles, the test level sets of the dispatched vehicles are summarized to obtain.

4. The driving assistance system performance evaluation system according to claim 3, characterized in that: The performance evaluation parameters of the auxiliary driving system of each dispatched vehicle are evaluated in the following specific methods: Obtaining the trajectory, speed data set, identification and avoidance data set and road condition data set of each driving from the driving parameters of each dispatched vehicle, and processing to obtain the avoidance performance evaluation parameters and conventional performance evaluation parameters of each dispatched vehicle in each driving; The speed data set includes speed at each time point; The identification and avoidance data set includes obstacle information, starting time point, ending time point and risk characteristic parameters of each avoidance; The road condition data set includes a road condition image set and a chassis-road distance set in each detection period; Acquire characteristic parameters of each meteorological data from the meteorological parameters of each dispatched vehicle, and obtain meteorological evaluation parameters of each dispatched vehicle at each driving time through data processing; The avoidance performance evaluation parameter ε of each dispatched vehicle in each driving _i , conventional performance evaluation parameter η _i and meteorological assessment parameter β _i Importing performance evaluation parameter models into driver assistance systems Output the performance evaluation parameter α(PI) of the auxiliary driving system of each dispatched vehicle _i , where ε′, η′, β′, ε _i→1 , η _i→1 They are respectively the ideal avoidance performance evaluation parameter, the ideal conventional performance evaluation parameter, the ideal meteorological evaluation parameter, the allowable deviation avoidance performance evaluation parameter corresponding to the unit deviation meteorological evaluation parameter in the data warehouse, and the allowable deviation conventional performance evaluation parameter.

5. The driving assistance system performance evaluation system according to claim 4, characterized in that: The specific evaluation method of the avoidance performance evaluation parameters of each dispatched vehicle in each driving is as follows: Extract the start time point, end time point and risk characteristic parameter F of each avoidance from the recognition avoidance data set of each dispatched vehicle. _imp , m is the number of each driving, m = 1, 2, ..., l, p is the number of each avoidance, p = 1, 2, ..., q; Based on the starting time point and the ending time point of each avoidance of each dispatched vehicle, combined with the trajectory of each dispatched vehicle, the trajectory of each avoidance of each dispatched vehicle is intercepted, and the speed data subset of each avoidance of each dispatched vehicle is intercepted. After data processing, the first avoidance performance evaluation parameter ε1 of each dispatched vehicle is output. _i ; Extract the road condition image set and the chassis-road distance set of each monitoring period from the road condition data set of each dispatched vehicle, and judge whether the road condition of each monitoring period of each dispatched vehicle is abnormal. If the road condition of a monitoring period is abnormal, the monitoring period is recorded as an abnormal monitoring period, and the abnormal monitoring periods of each dispatched vehicle are screened. Based on each abnormal monitoring period of each dispatched vehicle, a subset of trajectory and speed data corresponding to each abnormal monitoring period of each dispatched vehicle is intercepted, and after data processing, a second avoidance performance evaluation parameter ε2 of each dispatched vehicle is output. _i ; The first avoidance performance evaluation parameter and the second avoidance performance evaluation parameter of each dispatched vehicle are introduced into the avoidance performance evaluation parameter evaluation model ε _i =ln(1+ε1 _i *λ1+ε2 _i *λ2), the avoidance performance evaluation parameters of each dispatched vehicle for each driving are output, where λ1 and λ2 represent the weighted influence factors corresponding to the preset first avoidance performance evaluation parameter and the second avoidance performance evaluation parameter, respectively.

6. The driving assistance system performance evaluation system according to claim 5, characterized in that: The specific processing process of the first avoidance performance evaluation parameter of each driving of each dispatched vehicle is as follows: Through the avoidance trajectories of each dispatched vehicle, the convergence distance, trajectory length and avoidance duration of each dispatched vehicle are obtained, and the trajectory smoothness index of each avoidance of each dispatched vehicle is evaluated. In the formula It is represented as the jth vector of the pth avoidance of the mth travel of the i-th dispatched vehicle; Based on the obstacle information of each avoidance of each dispatched vehicle and the reference avoidance information stored in the data warehouse, wherein the reference avoidance information includes the basic information and avoidance trajectory of each obstacle, determine the reference convergence distance, reference trajectory smoothness index, trajectory length and avoidance duration of each avoidance of each dispatched vehicle; The speed dispersion index V of each avoidance of each dispatched vehicle during each travel is determined by using the speed data subset of each avoidance of each dispatched vehicle during each travel. _imp ; After the first avoidance performance evaluation parameter model Output the first avoidance performance evaluation parameter of each dispatched vehicle for each trip, χ _imph , χ′ _imph They are the h-th data and the h-th reference data of the p-th avoidance of the m-th travel of the ith dispatched vehicle, h∈[1,4].

7. The driving assistance system performance evaluation system according to claim 4, characterized in that: The specific evaluation method of the conventional performance evaluation parameters of each dispatched vehicle during each driving is as follows: Obtain each regular driving trajectory of each dispatched vehicle at each driving time, obtain the reference driving trajectory of each dispatched vehicle at each driving time from the data warehouse, and obtain the trajectory accuracy index Q_im of each dispatched vehicle at each driving time through trajectory fitting processing; Obtain a subset of the regular driving speed data of each dispatched vehicle in each section of each trip, and determine the speed dispersion index L of each dispatched vehicle in each section of each trip _imf , f is the number of each section of regular driving, f = 1, 2, ..., t; After routine performance evaluation parametric model Output the conventional performance evaluation parameters of each dispatched vehicle in each driving, where e is a natural constant.

8. The driving assistance system performance evaluation system according to claim 4, characterized in that: The specific evaluation method of the meteorological evaluation parameters of each dispatched vehicle during each driving is as follows: Import the characteristic parameters of each weather data of each trip in the weather parameters of each dispatched vehicle into the weather evaluation parameter model Output the meteorological assessment parameters of each dispatched vehicle at each driving time, where δ _imb is the characteristic parameter of the bth meteorological data of the mth driving in the meteorological parameters of the i-th dispatched vehicle, δ _ ' imb It is the characteristic parameter of the bth meteorological data in the ideal meteorological parameters stored in the data warehouse, b is the number of each meteorological data, b = 1, 2, ..., d.

9. The driver assistance system performance evaluation system according to claim 1, characterized in that: The specific method of performing the assisted driving system performance test on each dispatched vehicle is as follows: Based on the test level sets of dispatched vehicles, the interval step lengths of each test level in each variable test item are obtained from the data warehouse, the interval step lengths of each test level set of dispatched vehicles in each variable test item are screened, and the characteristic values ​​of each test level set of dispatched vehicles in each variable test item are set by using the control variable method; Based on the characteristic values ​​of each test level set of the dispatched vehicle in each test of each variable test item, the assisted driving system performance test is performed on each dispatched vehicle.

10. The driver assistance system performance evaluation system according to claim 8, characterized in that: The specific evaluation and judgment method for judging whether a dispatched vehicle needs to be recalled is as follows: The performance evaluation parameters of each test of each variable test item are obtained from the test data of the assisted driving system of each dispatched vehicle. If the performance evaluation parameter of a test is greater than or equal to the preset performance evaluation parameter threshold, the test is recorded as a successful test. The number of successful tests of each variable test item of the assisted driving system of each dispatched vehicle is counted. _ix And the total number of tests θ′ _ix , and import it into the test performance evaluation index model The test performance evaluation index of the assisted driving system of each dispatched vehicle is output, where θ _i0 is the number of successful tests of a certain variable test item of the i-th dispatched vehicle, τ′ _x is the successful test ratio of the xth independent variable item stored in the data warehouse, x is the number of each independent variable test item, x = 1, 2, ..., y; The test performance evaluation index includes values ​​of 0 and 1. When the test performance evaluation index is 0, it indicates that the test performance of the assisted driving system of the dispatched vehicle is unqualified and the dispatched vehicle needs to be recalled. When the test performance evaluation index is 1, it indicates that the test performance of the assisted driving system of the dispatched vehicle is qualified and the dispatched vehicle does not need to be recalled.

Images